Adaptive Sensing Reconfiguration for Cost-Effective Multi-Damage Monitoring of Civil Infrastructure

Infrastructure monitoring holds great promise in improving the nation's aging and deteriorating civil infrastructure. However, the monitoring needs to capture multiple classes of damage to attain its promise, because infrastructure often sustains multiple damages that occur simultaneously or sequentially. Although methods for multi-damage monitoring exist, they are not cost-effective, either requiring the costly deployment of multiple types of sensors or using a single type of sensor with a fixed sensing configuration. To address this knowledge gap, this paper aims to study the effectiveness of using different sensing configurations of a single sensor to monitor the same damage (cracking), thereby evaluating the need for and feasibility of adaptively reconfiguring a single sensor to monitor multiple damages. The research methodology includes finite element simulation to simulate and generate fiber optical sensor signals under various configurations and statistical testing to compare sensor signals generated under different sensing configurations. The preliminary experimental results show that the performance of the damage monitoring varied across sensor configurations: the cracking signals under configurations used to measure deflection, displacement, and velocity showed significant differences from non-damage signals. The results show the need for and feasibility of adaptive reconfiguration for multi-damage monitoring.